Natural Variation in Drug Susceptibility to HCV Polymerase Inhibitors in Treatment-naïve HCV Patient Isolates

From Journal of Viral Hepatitis

S.-C. C. Sun; A. Bae; X. Qi; J. Harris; K. A. Wong; M. D. Miller; H. Mo

Posted: 12/27/2011; J Viral Hepat. 2011;18(12):861-870. © 2011 Blackwell Publishing

Abstract and Introduction

Abstract

To assess the natural variation in drug susceptibility among treatment-naïve hepatitis C virus (HCV) patient isolates, the susceptibilities of chimeric replicons carrying the HCV NS5B polymerase from up to 51 patient isolates against a panel of diverse HCV nonnucleoside polymerase inhibitors were evaluated using a replicon-based transient replication assay. Some patient to patient variation in susceptibility to the panel of three HCV nonnucleoside polymerase inhibitors was observed. Linear regression and correlation analyses revealed no correlations among the susceptibilities to the polymerase inhibitors tested. Our results suggest that variable antiviral responses to HCV nonnucleoside polymerase inhibitors may be observed because of the natural variation in baseline susceptibility. In addition, the lack of correlation among the susceptibilities to three classes of HCV polymerase inhibitors evaluated here supports their possible combined use in a combination therapy strategy

Introduction

Hepatitis C virus (HCV) is a positive-strand RNA virus and is estimated to infect over 170 million people worldwide.^[1,2] To date pegylated interferon-α plus ribavirin, the current standard of therapy for HCV infection, is associated with incomplete efficacy and various side effects.^[3–6] Therefore, intense effort and time have been devoted to the discovery and development of novel and selective small-molecule inhibitors of HCV. Although no direct small-molecule antivirals have yet been approved for therapeutic use, the NS3 protease and NS5B polymerase are considered to be prime targets, and inhibitors of each enzyme have shown strong antiviral activity in early clinical trials.^[7–11]

HCV RNA-dependent RNA polymerase is the essential enzyme for replication of HCV RNA. A number of HCV polymerase inhibitors have been discovered; some have advanced to phase I/II clinical trials and have demonstrated antiviral activity in HCV-infected subjects in monotherapy.^[9,10,12] Among these HCV polymerase inhibitors, a number of nucleoside analogues (2'-Me-C, R1479, NM283, R1626 and R7128) bind to the active site of the HCV polymerase.^[13–15] In vitro resistance selection in the presence of nucleoside analogues identified a S282T resistance mutation for 2'-Me-C, and two (S96T and N142T) for R1479.^[16] The S282T mutation has also been reported in virologic breakthrough in patients on an NM283 (Idenix Pharmaceuticals, Cambridge, MA, USA) clinical trial.^[17,18] In addition to the active site, there are four allosteric binding sites for nonnucleoside inhibitors (NNIs) within the NS5B polymerase including: palm I (palm domain near the active site), palm II (partially overlapping palm I and towards the active site), thumb I (thumb domain near the fingertips) and thumb II (the outer surface of the thumb domain).^[19–21] First, the benzothiadiazine and acylpyrrolidine class of NNIs bind to the palm I domain of the polymerase.^[7,20] Resistance selection in the presence of benzothiadiazines revealed amino acid substitutions at residues 411, 414, 448, 451, 553, 554, 555, 556 and 559.^[22–25] Second, the benzofuran class of NNIs (HCV-796; Wyeth/Viropharma) binds to the palm II domain. HCV-796 (benzofuran) has been reported to select a major resistance mutation at amino acid residue 316 in an in vitro replicon system, and in early clinical trials.^[26,27] Distinct from the above HCV polymerase inhibitors, benzimidazoles and indoles bind to the thumb I site.^[28] Reduced susceptibility to this class of compounds was associated with the selection of mutations at residues 495 and 496 of the NS5B polymerase.^[28] Lastly, the thiophene carboxylic acids bind to the thumb II site.^[20,29,30] The most frequent mutations observed in vitro and in vivo for this class are located at amino acid position 423 (M423T/V/I) followed by positions 419 (L419M) and 482 (I482L).^[20,30,31]

HCV is characterized by a high degree of genetic diversity.^[32,33] The nucleotide sequences among the six different genotypes differ at 30–35% of the nucleotide sites.^[34,35] Each of the six major genotypes of HCV contains a series of more closely related subtypes that typically differ from each other by 20–25% in nucleotide sequences. Furthermore, 5–8% sequence divergence was observed between individual strains (variants) of HCV within a given subtype (GT-1a and GT-1b). It is possible that this high degree of natural sequence variation may have an impact on the susceptibility of patient isolates to HCV NS5B inhibitors, possibly, influencing the clinical response to direct antivirals. Therefore, it is of interest to test drug candidates against circulating genotypes from a variety of clinical isolates. In the present study, the susceptibilities of a panel of patient-derived NS5B polymerase to a number of nonnucleoside polymerase inhibitors were evaluated using a transient replication assay. The correlations in drug susceptibility among the different polymerase inhibitors were analysed.

Materials and Methods

Clinical Isolates

Patient serum samples were obtained from untreated GT-1a and GT-1b HCV-infected patients. All patients gave informed consent and originated from the USA.

Compounds

The NNIs benzofuran (NNI-1), benzothiadiazine (NNI-2) and thiophene carboxylic acid (NNI-3) were synthesized at Gilead Sciences, Inc. (Foster City, CA, USA).

Construction of NS5B Shuttle Vector

The genotype 1b-Con-1 subgenomic replicon construct 1b-PI-luc used to create the shuttle vectors has been described by Friebe et al.^[36] The components of the replicon are depicted in Fig. 1. A poliovirus IRES element was added at the 5' end after the HCV 5'NTR to increase firefly luciferase translation and RNA replication. Translation of HCV replicon from NS3 to NS5B is driven by the EMCV IRES. Three adaptive mutations, two in NS3 (E1202G+T1280I) and one in NS4B (K1846T) were introduced for efficient replication.^[37] Luciferase activity was used as an end-point readout.

Figure 1. Design of shuttle vector. Vector used for GT-1a and GT-1b NS5B gene insertion from the clinical isolates. The backbone of the shuttle vector is from 1b-Con-1.

To clone NS5B from patient isolates, a BclI site located at the 9th amino acid residue of NS5B was used as the 5'-end cloning site (Fig. 1). An AseI site was inserted directly after the TGA stop codon of NS5B, giving an insertion of ATTAAT at the 5' end of the 3' NTR. The restriction site was generated by site-directed mutagenesis using a QuickChange XL mutagenesis kit (Agilent, Santa Clara, CA, USA). The 1.0-kb fragment between two BglII sites in NS5B was removed to prevent contamination of recombinant shuttle vectors with the parental NS5B gene in luciferase readouts (Fig. 1).

NS5B Gene Isolations and Subcloning to Shuttle Vector

Viral RNA was isolated from 140 μL of serum from HCV-infected subjects using the QiaAmp Viral RNA isolation kit (Qiagen, Valencia, CA, USA) according to the supplier's instructions. cDNA was synthesized in a 20-μL reaction containing 0.2 mm of each dNTP, 0.125 μm of reverse primers and 50 U/μL MonsterScript reverse transcriptase (Epicentre, Madison, WI, USA) as recommended by the manufacturer. Ten microlitres of the RNA isolated from 140 μL of serum was denatured at 65 °C for 1 min, chilled on ice and added as template for reverse transcription. The reverse transcription reaction was incubated at 50 °C (1a) or 54 °C (1b) for 10 min and then at 60 °C for 40 min. Subsequently, the reverse transcriptase was heat inactivated at 90 °C for 2 min. The cDNA was used as template for PCR amplification of the NS5B gene.

A nested PCR strategy was used with genotype-specific primers to amplify the NS5B gene that was then used as a template for sequencing. GT-1a was amplified using primers 1aNS5B-5'7380-5' GAA TCA ACC CTA TCT ACT GCC TTG GCC GAG C-3' and 1aNS5B-3'9386-5' CTA AGA GGC CGG AGT GTT TAC-3' for the first round of PCR, and primers 1aNS5B-5'-7419-5'- TTT GGC AGC TCC TCA ACT TCC GG-3' and 1aNS5B-3'-9372-5' GAG TGT TTA CCC CAA CCT TCA TCG-3' for nested PCR. GT-1b was amplified using primers 1bNS5B-5'7595 -5' TCG TAC TCC TCC ATG CCC CCC CTT GA-3' and 1bNS5B-3'9459 -5' CCT ATT GGC CTG GAG TGT TTA GCT C-3' for the first round of PCR, and primers 1bNS5B-5'-7637-5'- GAT CTC AGC GAC GGG TCT TGG TC-3' and 1bNS5B-3'-9424-5' TTG GGG AGC AGG TAG ATG CCT AC-3' for nested PCR. PCR parameters were as follows for both genotypes: 94 °C, 2 min, 35 cycles at 94 °C for 30 s, 55 °C for 30 s, 72 °C for 2:10 min for first round PCR and 94 °C, 2 min, 35 cycles at 94 °C for 30 s, 60 °C for 30 s, 72 °C for 2:10 min for nested PCR. All nested PCRs were performed using the High Fidelity Platinum Taq DNA polymerase (Invitrogen, Carlsbad, CA, USA). PCR products were purified on a QiaQuick PCR clean-up column (Qiagen), according to the supplier's recommendations. The cycle sequencing of purified PCR product was conducted by ABI BigDye Terminator Sequencing technology (Applied Biosystems, Inc. Carlsbad, CA, USA) with 50 ng of purified PCR product. DNA sequencing conditions were 96 °C for 1 min, followed by 25 cycles of 96 °C for 10 s, 50 °C for 5 s and 60 °C for 4 min. The dried products were then denatured in Hi-Di formamide and run on an ABI 3100 genetic analyzer. DNA sequences were analysed with the Sequencher 4.0 program. Reference sequences of 1a-H77 and 1b-Con-1 were used as comparators to report any changes in the clinical isolates for GT-1a and GT-1b, respectively. Amino acid (AA) sequence analyses were performed from AA 1-591 and AA 1-582 of the NS5B gene for GT-1a and GT-1b, respectively.

The final PCR products for cloning into the NS5B shuttle vector were amplified from the above PCR products with primers containing 5' BclI and 3' AseI restriction sites. The final PCR was performed in the buffer supplied with the polymerase, plus 1.25 mm MgCl₂, 0.3 mm each primer. After digestion with AseI (New England Biolabs, Ipswich, MA, USA) at 37 °C for 3 h and subsequent BclI digestion at 50 °C for 3 h, the PCR product was cleaned with a QiaQuick PCR column (Qiagen) according to the supplier's recommendations. The shuttle vector DNA was similarly digested and gel purified with QIAEX II Gel Extraction kit (Qiagen) according to manufacturer's protocol. Purified shuttle vector DNA was then treated with shrimp alkaline phosphatase (Roche) for 15 min at 37 °C and then 65 °C for 15 min to heat inactivate the enzyme. Ligations were performed with 100 ng each of vector and insert DNA in 20 μL of 50 mM Tris–Cl, pH 7.8, 10 m MgCl2, 10 mM DTT, 1 mm ATP, 25 μg/mL bovine serum albumin and 0.15 Weiss units of T4 DNA ligase (New England Biolabs) at 16 °C, overnight.

Plasmid Purification and RNA Synthesis

The ligation reaction was coprecipitated with Pellet Paint® Co-Precipitant (Novagen, San Diego, CA, USA) according to the supplier's recommendations. The precipitated DNA was then resuspended in 2 μL of water and transfected by electroporation into Escherichia coli ElectroTen-Blue® Electroporation Competent Cells (Agilent) according to the supplier's recommendations. Ten per cent of the transformation mixture was plated on antibiotic selection plates to determine the transformation efficiency and the remaining transformants were expanded in liquid culture to propagate the quasi-species pool. Plasmid DNA from each overnight culture was then purified using two QiaPrep mini spin columns (Qiagen). Plasmid DNA was linearized by digestion with ScaI, and purified by extraction once with an equal volume of phenol/chloroform (1/1). DNA was precipitated with ethanol, and template RNA was synthesized using a T7 Megascript RNA synthesis kit (Ambion, Austin, TX, USA) according to the supplier's instructions. After lithium chloride precipitation, RNA was used for the transient replication assays.

Transient Replication Assay

The Huh-7 cells used in the transient replication assay were derived from a cured replicon cell clone as described previously,^[36] referred to as Lunet cells. For all experiments, Lunet cells were grown to a density of (6–10) × 10⁴ cells/cm² in Dulbecco's minimal essential medium (DMEM; Invitrogen) containing 10% foetal bovine serum (HyClone, Waltham, MA, USA), penicillin (100 U/mL)/streptomycin (100 μg/mL) (Gibco/Invitrogen) and 100 μM MEM Non-Essential Amino Acids Solution (Invitrogen). Cells were harvested by trypsinization and washed twice with ice-cold PBS; cell concentration was adjusted to 1 × 10⁷ cells/mL with ice-cold PBS, and 0.4 mL was transferred to a cold cuvette with a 0.4-cm gap, along with 5–10 μg of template RNA. Electroporation was performed with a Gene Pulser II (BioRad, Hercules, CA, USA) at 960 μF and 270 V using 1 manual pulse. Transfected cells were diluted to 2 × 10⁵ cells/mL and plated in 96-well plates at 2 × 10⁴ cells per well in complete DMEM medium. A duplicate plate was analysed for luciferase activity at 4 h posttransfection for transfection efficiency normalization. Compounds (or DMSO alone as a control) were added 24 h posttransfection in threefold dilutions at a final DMSO concentration of 0.5% (v/v). Firefly luciferase reporter signal was read 72 h after addition of compounds using the Luciferase assay system (Promega, Madison, WI, USA) with a Victor Luminometer (Perkin-Elmer, Waltham, MA, USA).

EC₅₀ Determination and Replication Capacity of HCV NS5B Clinical Isolates

The EC₅₀ values were assessed as the compound concentration at which a 50% reduction in the level of firefly reporter activity was observed when compared with control samples in the absence of compound. Dose response curves and EC₅₀ values were generated by using GraphPad Prism 4.0 (GraphPad Software, La Jolla, CA, USA) with nonlinear regression analysis. The signal-to-noise window was determined as the ratio of luciferase activity from cells treated with 0.5% DMSO vs activity from cells treated with 500 nM of BILN-2061 in 0.5% DMSO.

The replication level of either reference strains (1b-Con-1) or chimera replicons derived transiently from clinical isolates was determined as the ratio of the firefly luciferase signal at day 4 to that at 4 h postelectroporation, to normalize for transfection efficiency. The replication capacity of each chimera replicon derived from clinical isolates was expressed as their normalized replication efficiency when compared with that of the reference strain (1b-Con-1) within the same experiment.

Statistical Regression Analysis of Correlation Between Different HCV Inhibitors

The mean EC₅₀ values derived from each individual treatment-naïve HCV patient were compared in a statistical analysis to determine the correlation between different HCV inhibitors. This was calculated using the Pearson product-moment correlation coefficient (r) and the coefficient of determination (r²) of linear regression analysis. Both the correlation and linear regression analyses were performed using GraphPad Prism 4.0 (GraphPad Software).

Results

Characterization of the Phenotypic Assays

The plasmid used for the construction of the NS5B designated as 1b-PI-luc, includes three adaptive mutations (E1202G, T1280I and K1847T) and a poliovirus IRES element that has been shown to enhance both translation of the luciferase gene and RNA replication.^[36] This replicon replicated to high levels in Lunet cells. The NS5B shuttle vector was modified by addition of the 3' restriction site (AseI) for insertion of NS5B genes from clinical isolates. Insertion of the new restriction site did not adversely affect the replication capacity of the replicon (data not shown).

To evaluate the reproducibility of the NS5B phenotypic assay, the susceptibility of reference replicon (1b-Con-1) to the thiophene carboxylic acid, benzofuran and benzothiadiazine NNIs was obtained from between 15 and 16 independent determinants. Table 1 summarizes the mean EC₅₀ values, standard deviations, the 95% confidence intervals, and the maximum fold changes for the NNIs. The assay was highly reproducible with the EC₅₀ determinations from independent assays differing by <2-fold with respect to the mean EC₅₀ for each drug.

The Replication of the Chimeric Replicons Carrying Heterologous NS5B From Clinical Isolates

To test the clinical isolates, the HCV NS5B polymerase gene was amplified from 55 clinical isolates from untreated patients infected with HCV (41 GT-1a and 14 GT-1b) and were ligated into the NS5B shuttle vector. By mimicking the intrinsic HCV heterogeneity present in patients, these pooled populations of molecules were transfected into Lunet cells. The replication capacity of the patient isolates was determined by comparing the luciferase activity in the test sample to the parental 1b-PI-Luc at day 4 in a transient replication assay. As shown in Fig. 2, the replication capacity of the chimeric replicons harboring the NS5B genes from patient sera ranged from 0.01% to 50% of the wild-type 1b-Con-1. In this assay, the luciferase signal-to-noise ratio of the parental 1b-PI-Luc was high (>1000, data not shown). EC₅₀ values can be accurately determined if the clinical sample can replicate at 0.01% of the wild-type replicon (corresponding to signal/noise ratio of >10). Among the 55 patient isolates, 51 of 55 (92.7%) demonstrated sufficient replication levels to allow for drug susceptibility determination while the remaining four clinical isolates for NS5B phenotypic analysis either did not replicate or replicated at a level too low to allow for accurate EC₅₀ determination.

Figure 2. Relative replication capacity of chimeric NS5B clinical isolates from 52 HCV treatment-naïve patients.

Susceptibility of Treatment-naïve Clinical Isolates to HCV Polymerase Inhibitors

To investigate the natural variation in drug susceptibility of the baseline clinical isolates, the chimeric replicons were tested for their susceptibilities to three HCV NNIs. Susceptibilities to the thumb II inhibitor thiophene carboxylic acid were comparable to the laboratory strain 1b-Con-1 (Fig. 3a and Table 1). Thirty-nine of these samples were GT-1a and eight were GT-1b. Four of the clinical isolates failed to generate an EC₅₀ value because of poor curve fitting. The EC₅₀ values ranged from 100 to 764 nM with a mean EC₅₀ of 322 ± 148 nM for GT-1a and 344 ± 148 nM for GT-1b patients. Thus, no difference in susceptibility to the thiophene carboxylic acid tested was observed between GT-1a and GT-1b.

Against the benzofuran, the majority of the GT-1 isolates (45 of 51) were slightly more sensitive to the benzofuran than the laboratory stain 1b-Con-1 (Fig. 3b and Table 1). The EC₅₀ ranged from 1.3 to 13 nM and the mean EC₅₀ values were 5.9 ± 2.9 nM from 39 GT-1a patients and 5.8 ± 1.7 nM from 12 GT-1b patients, which were slightly lower than that of the laboratory strain 1b-Con-1 (9.74 nM. Similar to the thiophene carboxylic acid, no significant difference in sensitivity between GT-1a and GT-1b was observed for the benzofuran.

Figure 3. Susceptibility to HCV polymerase inhibitors of treatment-naïve HCV NS5B clinical isolates indicates as EC₅₀ values. (a) A total of 47 NS5B clinical isolates (39 GT-1a and 8 GT-1b) were tested to determine the sensitivity to thiophene. The mean EC₅₀ is 322 ± 148 nM from GT-1a patients and 344 ± 148 nM from GT-1b patients. (b) A total of 51 NS5B clinical isolates (39 GT-1a and 12 GT-1b) were tested to determine the sensitivity to benzofuran. The mean EC₅₀ is 5.9 ± 2.9 nM from GT-1a patients and 5.8 ± 1.7 nM from GT-1b patients. (c) A total of 44 NS5B clinical isolates (33 GT-1a and 11 GT-1b) were tested to determine the sensitivity to benzothiadiazine. The mean EC₅₀ is 302 ± 165 nM from GT-1a patients and 130 ± 44 nM from GT-1b patients.

In contrast, the GT-1a clinical isolates showed a decreased sensitivity to the palm I inhibitor benzothiadiazine compared to GT-1b clinical isolates, and the laboratory strain 1b-Con-1 (Fig. 3c and Table 1). The reduced susceptibility to the compound ranged from 1.1- to 3.4-fold in EC₅₀ values when compared to the mean of GT-1b clinical isolates. We were unable to obtain reportable EC₅₀ values for seven of the clinical isolates because of poor curve fitting. The mean EC₅₀ was 302 ± 165 nM from 33 GT-1a and 130 ± 44 nM from 11 GT-1b patients.

As shown in Fig. 3 and Table 2, some natural variation in susceptibility among these isolates was observed with all three different classes of HCV polymerase inhibitors tested (benzofuran, benzothiadiazine and thiophene carboxylic acid). The EC₅₀ values in the 95th percentile were 4.5-, 6- and 7.5-fold higher than the 5th percentile EC₅₀ values for the thiophene carboxylic acid, benzofuran, and benzothiadiazine compounds, respectively.

Genetic Diversity of Variants Within Clinical Isolates

Population sequence analysis of full-length NS5B was performed for all 55 baseline clinical isolates, and each clinical isolate was aligned against respective subtype reference, GT-1a: 1a H77-AF009606 or GT-1b: 1b-Con-1-AJ238799, to identify differences between patient and reference amino acid sequences. Genotypic analysis revealed intrinsic genetic diversity among the clinical isolates, but did not identify any known resistance mutations that have been reported to be associated with resistance to any of the three classes of compounds tested in 52 of 55 isolates including all 41 GT-1a isolates (Table 3). In contrast, 1 of 14 GT-1b isolates harboured S556S/G as a single mutant/wild-type mixture; S556S/G is known to confer resistance to the benzothiadiazines (Table 3). This isolate had EC₅₀ values of 546.5, 5.42 and 144.6 nM for the thiophene carboxylic acid, benzofuran and benzothiadiazine, respectively (Table 4). The EC₅₀ value against the benzothiadiazine was the third highest among the 11 GT-1b isolates, but was within the natural variation. In addition, 2 of 14 (14%) GT-1b isolates contained both C316N and S556G mutants, known to confer low-level of resistance to the benzofuran and/or benzothiadiazine classes (Table 3). The drug susceptibility data were available from only one of these two isolates because the chimeric replicons from the second replicated poorly. The EC₅₀ values were 9.14 nM for the benzofuran, the highest among the 12 GT-1b isolates tested, and 112.27 nM for the benzothiadiazine, comparable to other GT-1b isolates (Table 4).

Linear Regression Analysis of Correlation

To provide more information regarding susceptibilities of baseline clinical isolates to HCV nonnucleoside polymerase inhibitors, the susceptibilities of the baseline clinical isolates were compared pairwise among the three NNIs. Linear regression analysis was performed with the mean EC₅₀ values of each clinical isolate obtained in the transient phenotypic assays (Fig. 4). The correlations between the susceptibilities of the clinical isolates to any of the three NNIs tested were extremely poor as the coefficients of determination (r²) ranged between 0.0001 and 0.043. The P-values from the Pearson correlation of each analysis are much greater than the preset threshold value alpha of 0.05 (≥0.17). These data indicate there is no significant correlation among the susceptibilities to the HCV inhibitors tested in this study.

Figure 4. Correlation between different HCV inhibitors. The x-axis and y-axis represent mean EC₅₀ values from the corresponding NS5B chimeric clinical isolates. The line (–) represents the best unconstrained regression through the points. Sample size (n), the coefficient of determination (r²) and the P-value of each correlation plot are indicated.

Discussion

In this study, the sensitivity of up to 51 clinical baseline isolates from untreated HCV GT-1-infected patients to a panel of nonnucleoside HCV polymerase inhibitors was determined to investigate the effect of HCV genetic diversity on the inhibitor's antiviral potency. Some natural variation in susceptibility among these isolates was observed with all three different classes of HCV polymerase inhibitors tested (benzofuran, benzothiadiazine and thiophene carboxylic acid). The degree of the variation in EC₅₀ was slightly different among these three drug classes of nonnucleoside HCV polymerase inhibitors. The EC₅₀ values in the 95th percentile were 4.5-, 6.0- and 7.5-fold higher than the 5th percentile EC₅₀ values for the thiophene carboxylic acid, benzofuran and benzothiadiazine, respectively. Interestingly, no significant difference in sensitivity between GT-1a and GT-1b was observed for the tested benzofuran and thiophene carboxylic acid. In contrast, the benzothiadiazine was significantly more potent against GT-1b than GT-1a. The differential potency against GT-1a vs GT-1b for only the benzothiadiazine, and the high assay reproducibility for the reference standard suggests that the natural variation in drug susceptibility observed among baseline clinical isolates is not caused by assay variation. In addition, the results in the study are consistent with the findings of previous reports that described variable activity for HCV nonnucleoside polymerase inhibitors among clinical baseline isolates using similar phenotypic analysis assays.^[38,39]

Given the observation of the natural variation in drug susceptibility to these three classes of nonnucleoside inhibitors, it is possible that this natural variation may cause variable antiviral response among different patients. For example, one dose may be optimal for patients containing more sensitive variants, but this same dose may be suboptimal for patients who have less susceptible variants. Indeed, variable response to some of the HCV nonnucleoside polymerase inhibitors among different patients has been observed in clinical studies, especially in lower-dose groups. Furthermore, consistent with the in vitro findings of less potent activity against GT-1a than GT-1b for the benzothiadiazines in this study, the response in HCV GT-1a-infected patients was poorer than the GT-1b-infected patients during monotherapy with ABT-333 and ANA-598 (both benzothiadiazines).^[9,10] In addition to the susceptibility of baseline isolates, the pharmacokinetics of the drug also influences response. The natural variation in drug susceptibility coupled with pharmacokinetic information could be used to predict the response to drug treatment and provide guidance on the drug concentration needed for achieving a maximal response. If sufficiently high levels of drug exposure relative to the cluster of EC₅₀ values is achieved, the natural variation in drug susceptibility would be of less concern. Finally, the natural variation in baseline susceptibility could provide a threshold for defining abnormal reductions in drug susceptibility and serve as an indicator for an increased probability of drug resistance.

Sequence analysis of NS5B revealed C316N and S556G double mutants in 2 of the 14 GT-1b isolates. Phenotypic analysis of the one isolate with sufficient replication capacity to test drug susceptibility had the highest EC₅₀ against the benzofuran. This finding is consistent with previous reports, demonstrating that C316N confers a low-level of resistance to the benzofuran.^[26] However, the benzofuran susceptibility of this isolate was well within the natural variation of the GT-1a isolates and it had wild-type susceptibility to the benzothiadiazine tested in this study. Similarly, no significant change in EC₅₀ to the benzothiadiazine was seen in the isolate containing S556S/G. Previous studies showed that both C316N and S556G were associated with reduced susceptibility to other compounds of the benzothiadiazine class.^[23,24,26] The discrepancy may be due to the fact that the benzothiadiazine compound tested in this study has subtle differences in chemical structure with the benzothiadiazine compounds in previous studies resulting in a slightly different interaction with the NS5B polymerase.

In contrast to the above three GT-1b isolates, NS5B sequence analysis did not identify any of the mutations that are known to confer resistance to the tested compounds in 52 of 55 patient samples (41 GT-1a and 11 of 14 GT-1b). In addition, no clear pattern of the NS5B sequence was revealed in baseline clinical isolates with higher EC₅₀ values (data not shown). Thus, the natural variation in drug susceptibility observed in this study may be caused by the intrinsic genetic diversity of HCV.

The mean EC₅₀ values derived from each baseline clinical isolate against the benzofuran, benzothiadiazine and thiophene carboxylic acid were compared between drugs using the statistical regression analysis of correlation. Overall, the correlation was poor between any of the three compounds. These results suggest that the HCV variants which are less susceptible to one of these three classes of nonnucleoside polymerase inhibitors may not be less sensitive to the other two different classes of nonnucleoside polymerase inhibitors. Therefore, a combination of two or three different classes of nonnucleoside polymerase inhibitors may be advantageous for maximal antiviral response and reducing the selection of resistance. However, other aspects including synergistic/antagonistic inhibitory effects, drug–drug interactions and overlapping toxicity profiles should be taken into consideration for potential combination therapy strategies. The lack of correlation between two different classes of nonnucleotide polymerase inhibitors agrees with the fact that these three nonnucleoside inhibitors target different sites of the HCV polymerase and also exhibit different resistance profiles.

In summary, the activity of three classes of nonnucleoside polymerase inhibitors (benzofuran, benzothiadiazine and thiophene carboxylic acid) against a panel of baseline clinical isolates was evaluated using a replicon-based transient replication assay. Some variation in drug susceptibility was observed for all three classes of nonnucleoside polymerase inhibitors. However, there was no correlation between susceptibilities from one compound to another among those tested. Our findings suggest that the existing natural variation in baseline susceptibility should be taken into account for the optimal dose selection for the development of future nonnucleoside polymerase inhibitors. The lack of correlation between drug susceptibilities supports the combination of these different classes of HCV nonnucleotide polymerase inhibitors in the clinic.

References

1. Seeff LB. Natural history of chronic hepatitis C. Hepatology 2002; 36(5 Suppl. 1): 35–S46.
2. Purcell RH. Hepatitis C virus: historical perspective and current concepts. FEMS Microbiol Rev 1994; 14(3): 181–191.
3. Fried MW, Shiffman ML, Reddy KR et al. Peginterferon alfa-2a plus ribavirin for chronic hepatitis C virus infection. N Eng J Med 2002; 347(13): 975–982.
4. Fried MW. Side effects of therapy of hepatitis C and their management. Hepatology 2002; 36(Suppl. 1): S237–S244.
5. Foster GR. Past, present, and future hepatitis C treatments. Semin LiverDis 2004; 24(Suppl. 2): 97–104.
6. Foster GR. Review article: pegylated interferons: chemical and clinical differences. Aliment Pharmacol Ther 2004; 20(8): 825–830.
7. Kwong AD, McNair L, Jacobson I, George S. Recent progress in the development of selected hepatitis C virus NS3.4A protease and NS5B polymerase inhibitors. Current OpinPharmacol 2008; 8(5): 522–531.
8. Kieffer TL, Sarrazin C, Miller JS et al. Telaprevir and pegylated interferonalpha-2a inhibit wild-type and resistant genotype 1 hepatitis C virus replication in patients. Hepatology 2007; 46(3): 631–639.
9. Lawitz E, Rodriguez-Torres M, Cohen D, et al. Dose-dependent decrease in HCV viral load following two-day monotherapy with ABT-333 in treatment-naive, HCV genotype 1-infected subjects. 4th International Workshop on Hepatitis C resistance & new compounds, June, 25–26th, Boston, USA 2009; Abstract# 18.
10. Lawitz E, Rodriguez-Torres M, DeMicco M et al. Antiviral activity of ANA598, a potent non-nucleoside polymerase inhibitors, in chronic Hepatitis C patients, 44th Annual Meeting of the European Association for the Study of the Liver (EASL) in Copenhagen, Denmark, Absract# 1055. J Hepatol 2009; 50: S384.
11. Sarrazin C, Rouzier R, Wagner F et al. SCH 503034, a novel hepatitis C virus protease inhibitor, plus pegylated interferon alpha-2b for genotype 1 nonresponders. Gastroenterology 2007; 132(4): 1270–1278.
12. Klein CE, Cohen D, Menon R et al. Safety, tolerability and antiviral activity of the HCV polymerase inhibitor ABT-072 following single and multiple dosing in healthy adult volunteers and two days of dosing in treatment-naive HCV genotype-1-infected subjects. Dec 6–10, Kohala Coast, Hawaii, Abstract # 56. HepDART 2009; 5(Suppl. 1): 52–53.
13. Murakami E, Bao H, Ramesh M et al. Mechanism of activation of beta-D-2¢-deoxy-2¢-fluoro-2¢-c-methylcytidine and inhibition of hepatitis C virus NS5B RNA polymerase. AntimicrobAgents Chemother 2007; 51(2): 503–509.
14. Toniutto P, Fabris C, Bitetto D, Fumolo E, Fornasiere E, Pirisi M. R-1626, a specific oral NS5B polymerase inhibitor of hepatitis C virus. IDrugs 2008; 11(10): 738–749.
15. Brown NA. Progress towards improving antiviral therapy for hepatitis C with hepatitis C virus polymerase inhibitors. Part I: nucleoside analogues. Expert Opin Investig Drugs 2009; 18(6): 709–725.
16. Ali S, Leveque V, Le Pogam S et al. Selected replicon variants with low-level in vitro resistance to the hepatitis C virus NS5B polymerase inhibitor PSI-6130 lack cross-resistance with R1479. Antimicrob AgentsChemother 2008; 52(12): 4356–4369.
17. O_Brien C, Godofsky E, Rodriguez-Torres M, et al. Randomized trial of Valopicitabine (NM283), alone or with Peg-Interferon, vs. retreatment with Peg-Interferon plus Ribavirin (PEGIFN/RBV) in hepatitis C patients with previous non-response to PEGIFN/RBV: first interim results. 56th Annual Meeting of AASLD Hepatology 42 (S1), 234A (Abstract 96). 2005.
18. Afdhal N, Godofsky E, Dienstag J, et al. Final phase I/II trial results for NM283, a new polymerase inhibitor for hepatitis C: antiviral efficacy and tolerance in patients with HCV-1 infection, including previous interferon failures. 55th Annual Meeting of AASLD Hepatology 40 (S4), 726A (Abstract LB03). 2004.
19. Love RA, Parge HE, Yu X et al. Crystallographic identification of a noncompetitive inhibitor binding site on the hepatitis C virus NS5B RNA polymerase enzyme. J Virol 2003; 77(13): 7575–7581.
20. Beaulieu PL. Non-nucleoside inhibitors of the HCV NS5B polymerase: progress in the discovery and development of novel agents for the treatment of HCV infections. CurrOpin Investig Drugs 2007; 8(8): 614–634.
21. Sarisky RT. Non-nucleoside inhibitors of the HCV polymerase. J AntimicrobChemother 2004; 54(1): 14–16.
22. Wagner R, Maring C, Flentge P, et al. Preclinical characterization of ABT-333 and ABT-072: novel nonnucleoside HCV NS5B polymerase inhibitors. Dec 6–10, Kohala Coast, Hawaii, Abstract # 108. Hep DART 2009; 5(Suppl. 1):100–101.
23. Lu L, Dekhtyar T, Masse S et al. Identification and characterization of mutations conferring resistance to an HCV RNA-dependent RNA polymerase inhibitor in vitro. AntiviralRes 2007; 76(1): 93–97.
24. Mo H, Lu L, Pilot-Matias T et al. Mutations conferring resistance to a hepatitis C virus (HCV) RNA-dependent RNA polymerase inhibitor alone or in combination with an HCV serine protease inhibitor in vitro. AntimicrobAgents Chemother 2005; 49(10): 4305–4314.
25. Showalter RE, Thompson PA, Steffy KR, Appleman JR. ANA598 displays potent in vitro antiviral activity against diverse clinical isolates of genotype 1 HCV in a transient replicon shuttle vector system. American Association for the Study of Liver Diseases (AASLD), Boston, Oct 30–Nov. 3, Abstract# 1586. Hepatology 2009; 50(Suppl. 4): 1037A.
26. Kneteman NM, Howe AY, Gao T et al. HCV796: a selective nonstructural protein 5B polymerase inhibitor with potent anti-hepatitis C virus activity in vitro, in mice with chimeric human livers, and in humans infected with hepatitis C virus. Hepatology 2009; 49(3): 745–752.
27. Flint M, Mullen S, Deatly AM et al. Selection and characterization of hepatitis C virus replicons dually resistant to the polymerase and protease inhibitors HCV-796 and boceprevir (SCH 503034). AntimicrobAgents Chemother 2009; 53(2): 401–411.
28. Kukolj G, McGibbon GA, McKercher G et al. Binding site characterization and resistance to a class of nonnucleoside inhibitors of the hepatitis C virus NS5B polymerase. J Biol Chem 2005; 280(47): 39260–39267.
29. Cooper C, Lawitz EJ, Ghali P et al. Evaluation of VCH-759 monotherapy in hepatitis C infection. J Hepatol 2009; 51(1): 39–46.
30. Shi ST, Herlihy KJ, Graham JP et al. In vitro resistance study of AG-021541, a novel onnucleoside inhibitor of the hepatitis C virus RNA-dependent RNA polymerase. Antimicrob Agents Chemother 2008; 52(2): 675–683.
31. Shi ST, Herlihy KJ, Graham JP et al. Preclinical characterization of PF-00868554, a potent nonnucleoside inhibitor of the hepatitis C virus RNA-dependent RNA polymerase. Antimicrob Agents Chemother 2009; 53(6): 2544–2552.
32. Herring BL, Tsui R, Peddada L, Busch M, Delwart EL. Wide range of quasispecies diversity during primary hepatitis C virus infection. J Virol 2005; 79(7): 4340–4346.
33. Zhang YY, Lok AS, Chan DT, Widell A. Greater diversity of hepatitis C virus genotypes found in Hong Kong than in mainland China. J ClinMicrobiol 1995; 33(11): 2931–2934.
34. Simmonds P. Viral heterogeneity of the hepatitis C virus. J Hepatol 1999; 31(Suppl. 1): 54–60.
35. Simmonds P. The origin and evolution of hepatitis viruses in humans. J Gen Virol 2001; 82(Pt 4): 693–712.
36. Friebe P, Lohmann V, Krieger N, Bartenschlager R. Sequences in the 5¢ nontranslated region of hepatitis C virus required for RNA replication. J Virol 2001; 75(24): 12047–12057.
37. Lohmann V, Hoffmann S, Herian U, Penin F, Bartenschlager R. Viral and cellular determinants of hepatitis C virus RNA replication in cell culture. J Virol 2003; 77(5): 3007–3019.
38. Middleton T, He Y, Pilot-Matias T et al. A replicon-based shuttle vector system for assessing the phenotype of HCV NS5B polymerase genes isolated from patient populations. J VirolMethods 2007; 145(2): 137–145.
39. Le Pogam S, Seshaadri A, Kosaka A et al. Existence of hepatitis C virus NS5B variants naturally resistant to non-nucleoside, but not to nucleoside, polymerase inhibitors among untreated patients. J Antimicrob Chemother 2008; 61(6): 1205–1216.

Source